Gapping was introduced by Ross (1967) as a conjunction reduction rule that deletes the repeated verbs in coordinate structures. See the following example from English where the application of gapping on (1a) results in (1b).
(1) Gapping (borrowed from Ross 1967):
(a) The boy works in a skyscraper and the girl works in a quonset hut.
(b) The boy works in a skyscraper and the girl in a quonset hut.
Some languages have "forward gapping" as shown in English (1) above, the common verb in the second conjunct clause is deleted. However, some languages have "backward gapping", thus the common verb from the first conjunct clause may also be deleted.
Gapping is assumed to have properties, such as (i) it needs to have lexical material on both its sides, it must occur in a coordinate structure, it may not be a phrase, it may not occur at the sentence boundary or violate complex NP constraint (Lobeck 1995), the gapped elements must be contextually given and the remnants must occur in a contrastive relation to their correlates (Winkler 1997 as mentioned in Winkler 2005, Johnson 1996).
There are two competing analyses for the mechanism of the gapping phenomenon, viz. the deletion of the material, and the across-the board movement of the material from both the conjuncts, these both achieve the similar results.
1. Johnson, K. (1996) In search of the middle field. ms. Accessed at http://people.umass.edu/kbj/homepage/index_johnson.htm.
2. Lobeck, A. (1995) Ellipsis. Oxford University Press.
3. Neijt, A. (1979) Gapping: A Contribution to Sentence Grammar. Foris Publications. Dordrecht.
4. Ross, J.R. (1967) Constraints on Variables in Syntax. PhD Dissertation. MIT.